COVID-19: A Comprehensive Review of Learning Models

Area of Science:

• Medical Informatics
• Computational Biology
• Epidemiology

Background:

• Coronavirus disease (COVID-19) is a contagious illness impacting the immune system, affecting numerous countries globally.
• Extensive research is ongoing to manage the outbreak through data analysis and monitoring.
• This paper reviews studies focused on early symptom detection, pandemic end-point estimation, and analysis of public discourse.

Purpose of the Study:

• To provide a comparative overview of research utilizing machine learning and deep learning for COVID-19 analysis.
• To summarize findings, limitations, and future research directions in this field.

Main Methods:

• Review of scientific literature on COVID-19 research.
• Analysis of studies employing machine learning and deep learning techniques.
• Examination of datasets including Chest X-rays, CT scans, and social media data (tweets).

Main Results:

• Machine learning and deep learning models like K-means, Random Forest, CNN, LSTM, Auto-Encoder, and Regression were applied to various datasets.
• Studies explored early symptom identification, pandemic trajectory estimation, and public sentiment analysis.
• Results and limitations of these computational approaches are detailed.

Conclusions:

• Machine learning and deep learning offer powerful tools for analyzing diverse COVID-19 data.
• Further research is needed to address existing challenges and refine these models for pandemic response.
• The review highlights the critical role of computational methods in understanding and managing infectious disease outbreaks.